[Proj] Lat-Long to Lambert Conformal Conic(2SP) transformation inaccuracy issue

[Tonu Lukk]

> In this case, the precision in the conversion from DMS to fractional degree 
> causes the slight error.  From gp

Makes sense. Yet I still can't figure out why the +lat_0=57d31\'03.19415\"N
gets interpereted wrong so that I end up with northing value approx. 98 
meters off the correct value. When I substitute the +lat_0 DMS value 
with its decimal equivalent everything works just fine and I get the 
expected northing - 6375000.

Tonu